strided-einsum2 (binary einsum)
The strided-einsum2 crate provides einsum2_into for binary tensor contractions.
Most Rust benchmark runners were moved to strided-opteinsum/benches/.
Latest broad benchmark results are documented in strided-opteinsum/README.md.
The dot_general batched matmul diagnostic compares the tenferro-benchmark
bij,bjk->bik cases against PyTorch bmm with memory-matched row-major and
col-major layouts. Allocation and setup are outside the timed loop: Rust
prepares GEMM operands once, and PyTorch uses torch.bmm(..., out=...).
On macOS, the runner uses blas-accelerate by default and verifies the
benchmark binary with otool -L. Losing this Accelerate path is a benchmark
regression. Use STRIDED_EINSUM2_DOT_GENERAL_RUST_FEATURES to test another
provider such as parallel,blas-openblas or parallel,blas-mkl.
Set STRIDED_EINSUM2_DOT_GENERAL_BENCH_DIAGNOSTICS=1 to emit extra Rust-side
diagnostic rows (raw-cblas-dgemm, raw-trait-dgemm, and on macOS
raw-fortran-dgemm) for separating BLAS call overhead from PyTorch bmm.
On macOS without MKL, PyTorch CPU bmm falls back to per-batch addmm rather
than MKL batched GEMM; current 1T Accelerate runs do not show a stable TN-layout
loss once measured with enough repetitions.
STRIDED_EINSUM2_DOT_GENERAL_BENCH_PROFILE=full \
STRIDED_EINSUM2_DOT_GENERAL_BENCH_DTYPES=f64,c64,c128 \
Latest local allocation-free Accelerate run (runs=30, warmups=5):
| benchmark | dtype | threads | strided-einsum2-accelerate-prepared ms | pytorch-bmm ms | ratio |
|---|---|---|---|---|---|
bin_batched_matmul_b32_m64_n64_k64 |
f64 | 1 | 0.046375 | 0.077375 | 0.60x |
bin_batched_matmul_b32_m64_n64_k64 |
f64 | 4 | 0.046417 | 0.078521 | 0.59x |
bin_batched_matmul_b32_m64_n64_k64 |
c64 | 1 | 0.343666 | 0.332021 | 1.04x |
bin_batched_matmul_b32_m64_n64_k64 |
c64 | 4 | 0.342417 | 0.332604 | 1.03x |
bin_batched_matmul_b32_m64_n64_k64 |
c128 | 1 | 0.748334 | 0.760542 | 0.98x |
bin_batched_matmul_b32_m64_n64_k64 |
c128 | 4 | 0.755875 | 0.758333 | 1.00x |
bin_batched_matmul_b32_m128_n128_k128 |
f64 | 1 | 0.301042 | 0.376313 | 0.80x |
bin_batched_matmul_b32_m128_n128_k128 |
f64 | 4 | 0.307916 | 0.378687 | 0.81x |
bin_batched_matmul_b32_m128_n128_k128 |
c64 | 1 | 1.224958 | 1.251854 | 0.98x |
bin_batched_matmul_b32_m128_n128_k128 |
c64 | 4 | 1.238375 | 1.243312 | 1.00x |
bin_batched_matmul_b32_m128_n128_k128 |
c128 | 1 | 2.981833 | 2.979667 | 1.00x |
bin_batched_matmul_b32_m128_n128_k128 |
c128 | 4 | 2.999375 | 2.977917 | 1.01x |
The faer-prepared path is useful for isolating non-BLAS behavior, but macOS regression checks should use the Accelerate path above.
Julia reference scripts (e.g. julia_matmul.jl, julia_dot.jl) use OMEinsum. Run single-threaded for comparison (from repo root):
OMP_NUM_THREADS=1 JULIA_NUM_THREADS=1
Example: julia_matmul.jl, julia_dot.jl, julia_trace.jl, julia_tcontract.jl, julia_outer.jl, etc.